in progress

Draft Research Plan

Colorectal Cancer: Screening

January 03, 2019

Recommendations made by the USPSTF are independent of the U.S. government. They should not be construed as an official position of the Agency for Healthcare Research and Quality or the U.S. Department of Health and Human Services.

Figure 1 is the analytic framework that depicts the three Key Questions to be addressed in the systematic review. The figure illustrates how screening for colorectal cancer in adults age 40 years or older may result in a decrease in colorectal cancer indicidence, colorectal cancer mortality, and/or all-cause mortality (Key Question 1). There is also a question related to the accuracy of screening tests used to detect colorectal cancer or adenomatous polyps (Key Question 2), and the potential harms of screening (Key Question 3).

Abbreviations: CTC = computed tomography colonography; FIT = fecal immunochemical test; FIT-DNA = fecal immunochemical test plus stool DNA test; FS = flexible sigmoidoscopy; gFOBT = guaiac-based fecal occult blood test; mSEPT9 = methylated septin 9 gene DNA.

  1. What is the effectiveness or comparative effectiveness of screening programs in reducing colorectal cancer, mortality, or both?
    1. Does the effectiveness of screening programs vary by age, sex, or race/ethnicity?
  2. What is the accuracy of direct visualization, stool-based, or blood-based screening tests for detecting colorectal cancer, advanced adenomas, or adenomatous polyps based on size?
    1. Does the accuracy of the screening tests vary by age, sex, or race/ethnicity?
  3. What are the serious harms of the different screening tests?
    1. Do the serious harms of the screening tests vary by age, sex, or race/ethnicity?

Contextual questions will not be systematically reviewed and are not shown in the Analytic Framework.

  1. What is the adherence (or uptake) to testing for each of the currently available screening tests?
  2. What is the adherence to followup diagnostic colonoscopy for abnormal screening test results (i.e., from stool-based testing, flexible sigmoidoscopy, or computed tomography colonography)?
  3. Does the natural history of adenomas vary by age, sex, or race/ethnicity?
  4. Does the prevalence or distribution of colorectal lesions (colorectal cancer, advanced adenomas, or small adenomatous polyps) in the colon vary by age, sex, or race/ethnicity?
  5. What models or tools are available for assessing the risk of developing colorectal cancer in average-risk screening populations?
  6. How are newer screening tests (e.g., fecal immunochemical test plus stool DNA test, computed tomography colonography) being implemented in clinical care, and what are the implementation concerns around these tests?
  7. Are there colorectal cancer screening decision models, other than CISNET (Cancer Intervention and Surveillance Modeling Network), that address the age to start screening, age to stop screening, and comparative effectiveness of various screening strategies? What are their findings?

The Proposed Research Approach identifies the study characteristics and criteria that the Evidence-based Practice Center will use to search for publications and to determine whether identified studies should be included or excluded from the Evidence Report. Criteria are overarching as well as specific to each of the key questions (KQs).

  Include Exclude
Populations Adults age ≥40 years in average-risk or unselected populations; screening populations (i.e., asymptomatic persons) Populations selected for personal or family history of colorectal cancer (one or more first-degree relatives with colorectal cancer diagnosed before age 60 years or two or more first-degree relatives with colorectal cancer diagnosed at any age), a known genetic susceptibility syndrome (e.g., Lynch syndrome, familial adenomatous polyposis), or personal history of inflammatory bowel disease; nonscreening populations (e.g., persons who are symptomatic, test positive on screening, have iron deficiency anemia, or are under surveillance for previous colorectal lesion)
Settings Settings representative of community practice for flexible sigmoidoscopy and colonoscopy studies; studies conducted in countries categorized as “very high” on the 2017 Human Development Index (as defined by the United Nations Development Programme) Primarily research-based settings (or select academic settings that would not be applicable to most practice settings) for endoscopy studies (e.g., small studies aimed at evaluating new endoscopy technologies, studies with operator or resource characteristics not applicable to community practice); studies conducted in countries not categorized as “very high” on the 2017 Human Development Index
Screening tests KQ 1: Any program of colorectal cancer screening, including endoscopy, imaging, stool-based, or blood-based testing

KQs 2, 3: Direct visualization tests:

  • Colonoscopy
  • Flexible sigmoidoscopy
  • Computed tomography colonography
  • Capsule endoscopy (PillCam®)

Stool-based tests:

  • High-sensitivity guaiac-based fecal occult blood test (gFOBT) (i.e., Hemoccult SENSA®)
  • Fecal immunochemical test (FIT) (quantitative and qualitative testing)
  • Fecal immunochemical test plus stool DNA test (FIT-DNA)

Blood-based test:

  • Circulating methylated septin 9 gene DNA test (mSEPT9)
KQs 2, 3: Hemoccult II (review of test performance and harms limited to high-sensitivity gFOBT); stool-based testing using in-office digital rectal examination; double contrast barium enema; magnetic resonance colonography
Comparisons KQ 1: No screening or alternate screening strategy

KQ 2: Diagnostic accuracy studies must use colonoscopy as a reference standard

KQ 3: No comparator necessary


Outcomes KQ 1: Colorectal cancer incidence (by stage), interval colorectal cancer; colorectal cancer–specific or all-cause mortality

KQ 2: Test accuracy, including: sensitivity and specificity (per person for all tests and per lesion for direct visualization); positive and negative predictive value (per person for all tests and per lesion for direct visualization); false-positive and false-negative rates; for colorectal cancer, advanced adenoma (high-grade dysplasia, villous histology, and/or size ≥10 mm) and/or adenomatous polyps by size (i.e., ≤5 mm, 6 to 9 mm, ≥10 mm) or by location in colon (e.g., proximal vs. distal)

KQ 3: Serious harms requiring unexpected or unwanted medical attention and/or resulting in death (e.g., requiring hospitalization), including but not limited to: perforation, major bleeding, severe abdominal symptoms, and cardiovascular events; extracolonic findings, subsequent diagnostic workup, and adverse events from diagnostic testing for incidental findings on computed tomography colonography; radiation exposure per computed tomography colonography examination

KQ 1: Incidence of adenomas or advanced neoplasia (composite outcome of advanced adenomas and colorectal cancer)

KQ 3: Minor harms defined as those not necessarily needing or resulting in medical attention (e.g., patient dissatisfaction, anxiety/worry, minor gastrointestinal complaints)

Study design All KQs: Fair- and good-quality studies

KQ 1: Randomized, controlled trials; controlled clinical trials; and prospective cohort studies

KQ 2: Randomized, controlled trials; controlled clinical trials; cohort studies; nested case-control diagnostic accuracy studies; and screening registry studies

KQ 3: Randomized, controlled trials; controlled clinical trials; large screening registry or database observational studies; cohort studies; and systematically selected case series

All KQs: Poor-quality studies

KQ 1: Decision analyses*

KQ 2: Diagnostic accuracy studies without a reference standard or without representation of a full spectrum of disease (e.g., case-control studies, excluded indeterminate results)

KQ 3: Case studies

* This review will be accompanied by commissioned collaborative microsimulation decision analyses from CISNET.

In addition to this updated systematic review, the USPSTF has commissioned a set of decision analyses from CISNET to support the recommendation process, as it did for the previous updates in 2008 and 2016 (profiles of the CISNET models can be found at: The commissioned decision analyses will help inform how the benefits and harms of colorectal cancer screening strategies may vary by the type of screening test, different starting and stopping ages, and the time interval between repeated screenings.